Changes in the pattern of bile acids in the nuclei of rat liver cells during hepatocarcinogenesis.

Bile acids reach the nuclei of hepatocytes, where they may play an important role in controlling gene expression by binding to nuclear receptors. In previous studies, changes in the amounts of the different molecular species of bile acids in the hepatocyte nucleus during rat liver regeneration have been reported. The aim of the present work was to investigate whether this also occurs during rat hepatocarcinogenesis. Liver cell nuclei were isolated after homogenization of livers from healthy adult rats (controls) and from rats at different time points during chemically induced hepatocarcinogenesis, corresponding to the stages of foci (12 weeks), hepatoma (20 weeks) and carcinoma (32 weeks). Bile samples from the cannulated common bile duct were collected for 1h from different sets of animals undergoing hepatocarcinogenesis. Bile acids in bile, liver homogenates and isolated nuclei were measured by GC-MS. Because the yield of nuclei isolated changed during the course of hepatocarcinogenesis (control, 20.1%; 12 weeks, 23.6%; 20 weeks, 7.8%; 32 weeks, 5.1%), amounts of bile acids in nuclei were corrected for the amount of DNA in each preparation. During hepatocarcinogenesis, bile acid concentrations in liver homogenates were reduced to approximately half the values obtained in control livers, while the levels of bile acids in both isolated nuclei and bile were not decreased. Hepatocarcinogenesis induced changes in the composition of bile acid pools. These were manifest as an increase in the proportion of cholic acid and a decrease in that of ursodeoxycholic acid in both bile and liver. These modifications differed from the changes seen in the nuclear bile acid pool, where a decrease in the proportion of cholic acid together with an increase in that of ursodeoxycholic acid were the major changes observed during hepatocarcinogenesis. With regard to the 'flat' bile acids (allo-cholic acid plus Delta(5)- or Delta(4)-unsaturated bile acids), a marked hepatocarcinogenesis-induced increase in the output of these species in bile was found. However, these bile acids were only found in liver homogenates at the hepatoma stage, whereas they were not detected in isolated nuclei at any stage of hepatocarcinogenesis. In summary, these results support the existence of a bile acid pool in hepatocyte nuclei whose composition differs from that of the extranuclear bile acid pool. Moreover, they indicate that, during hepatocarcinogenesis, the composition of the nuclear pool undergoes important alterations.

Predominance of human versus rat phenotype in the metabolic pathways for bile acid synthesis by hybrid WIF-B9 cells.

The rat hepatoma-human fibroblast hybrid cell line WIF-B9 stably exhibits the structural and functional characteristics of normal differentiated hepatocytes. The abilities of these cells to synthesize bile acids and amidate them with glycine and taurine were investigated. The release of bile acids into the culture media over 72 h was assessed by gas chromatography-mass spectrometry. WIF-B9 cells were able to synthesize bile acids (1.10+/-0.17 nmol/mg protein) but less efficiently than rat hepatocytes in primary culture (2.19+/-0.19 nmol/mg protein; P<0.01). The patterns of major bile acid species produced by both types of cells were also different. Cholic acid (CA; 72%) and beta-muricholic acid (19%) were the major bile acids produced by rat hepatocytes, while chenodeoxycholic acid (CDCA) accounted for only 4.5% of total bile acids. In contrast, muricholic acids were absent, while CA (62%) and CDCA (34%) were the most abundant bile acids synthesized by WIF-B9 cells. Using reverse transcription-polymerase chain reaction and gene- and species-specific primers for key enzymes involved in bile acid synthesis, the expression of human, but not rat, orthologues of CYP7A1, CYP27, CYP8B and CYP7B1 was found in WIF-B9 cells. Induction of cell stress by serum deprivation did not change the amount of total bile acids synthesized by these cells, but an inversion of the CA-to-CDCA ratio from 1.8 to 0.3 together with a marked increase in the proportion of intermediate metabolites related to the acidic pathway was found. Using 500 microM radiolabeled CA and 2 mM of taurine or glycine, the ability to amidate CA over 48 h was determined by high performance liquid chromatography. Rat hepatocytes conjugated more than 90% CA with either amino acid, whereas this ability was very poor (< 2%) in WIF-B9 cells. Regarding the expression of enzymes and the products of bile acid synthesis, it may be concluded that the human phenotype predominates over that of the rat in WIF-B9 cells. Moreover, these cells are almost completely unable to further conjugate primary bile acids, which facilitates the manipulation of these steroids in analytical procedures. These characteristics make WIF-B9 cells a suitable in vitro model to carry out studies on bile acid synthesis by 'human-like' metabolic pathways.

Increased levels of typically fetal bile acid species in patients with hepatocellular carcinoma.

The aim of this work was to investigate the reappearance during liver neoplasia of bile acids (BAs) species, which are unusual in healthy adults, but common in fetuses. Serum and urine samples were collected from patients with hepatocellular carcinoma (HCC; n=27), and for comparative purposes, with liver cirrhosis (n=49), liver metastasis (n=19), chronic viral hepatitis (n=11) and healthy volunteer (control group; n=26) groups. BAs were identified and measured by GC--MS. Hypercholanaemia was found in all groups of patients. In HCC, this was characterized by a marked increase in the chenodeoxycholate/cholate ratio in both serum and urine. Although increased levels of BAs, with hydroxylations at unusual positions, and oxo-BAs were found in HCC, these were not significantly different from those observed in other groups. However, BAs with a flat structure, i.e. Delta(4)-unsaturated- and 5 alpha- or allo-BAs, which were almost absent in healthy subjects, were markedly increased in the serum and urine of HCC patients. They were also detected, although in much lower amounts, in liver metastasis and liver cirrhosis, but not in viral hepatitis. Flat-BAs were better detected in urine than in serum. Urinary Delta(4)-unsaturated-BA output was significantly lower in patients with small tumours (<3 cm) compared with those with higher size tumours. No correlation between flat-BA output into urine and serum alpha-fetoprotein or total BAs was found. These results suggest that Delta(4)- and/or allo-BAs are particularly elevated in patients with HCC, which may be a potentially useful complementary, rather than alternative, marker for early detection of liver neoplasia.